Accounting method and device for greenhouse gas emission amount of lead smelting plant and storage medium

Abstract

The invention discloses an accounting method and device for greenhouse gas emission of a lead smelting plant and a storage medium, relates to the technical field of gas analysis, and solves the problems that in the prior art, when tail gas at different stages is subjected to accounting, when faults of a data sensor are not found in time, the accounting result is inaccurate, and the calculation time is shortened. The technical problem that excessive greenhouse gas is discharged is solved. Raw material data of a lead smelting plant are obtained; tail gas in the lead smelting process is detected through a data sensor, and gas content data is obtained; predicting the gas content in the tail gas according to the raw material data to obtain predicted content data; carrying out fusion analysis on the gas content data to obtain tail gas content data; comparing the predicted content data with the tail gas content data to obtain a predicted feedback coefficient; adjusting prediction according to the prediction feedback coefficient; the accounting accuracy can be improved.

Description

TECHNICAL FIELD

[0001] The application belongs to the field of gas analysis, and relates to a greenhouse gas emission accounting technology, in particular to a method and device for accounting for greenhouse gas emissions of a lead smelting plant and a storage medium. BACKGROUND

[0002] The greenhouse gas of the lead smelting plant refers to a gas that can absorb and re-radiate infrared rays, resulting in atmospheric warming, which is directly or indirectly produced in the production process; a large amount of greenhouse gas emissions will increase the greenhouse effect, and a large amount of lead-containing smoke and aerosols will be produced in the high-temperature smelting process of the lead smelting plant, which will be discharged with waste gas and settled around the plant, and will enter the human body through breathing or hand-mouth contact, causing lead poisoning; by accounting for the lead smelting plant, the source of greenhouse gas emissions can be determined, so that targeted emission reduction measures can be developed according to energy consumption, raw material decomposition and other factors in the smelting process, and the greenhouse gas emission can be accounted for regularly to monitor the implementation effect of the emission reduction measures, timely adjust the strategy and ensure the realization of the emission reduction target, thereby reducing the harm of greenhouse gas to the environment.

[0003] The prior art generally accounts for the greenhouse gas emissions by monitoring the smelting production process in real time, collecting tail gas emission data at different stages, and statistically accounting for the greenhouse gas in the tail gas; however, in the prior art, the tail gas at different stages is accounted for, and when the fault of the data sensor is not found in time, the result of the accounting is inaccurate, causing excessive greenhouse gas emissions. SUMMARY

[0004] The present application aims to at least solve one of the technical problems existing in the prior art; for this purpose, the present application provides a method and device for accounting for greenhouse gas emissions of a lead smelting plant and a storage medium, which are used to solve the technical problem that in the prior art, the tail gas at different stages is accounted for, and when the fault of the data sensor is not found in time, the result of the accounting is inaccurate, causing excessive greenhouse gas emissions.

[0005] To achieve the above-mentioned purpose, the first aspect of the present application provides a method for accounting for greenhouse gas emissions of a lead smelting plant, comprising: obtaining raw material data of the lead smelting plant; detecting the tail gas in the lead smelting process by a data sensor to obtain gas content data; predicting the gas content in the tail gas according to the raw material data to obtain predicted content data; performing fusion analysis on the gas content data to obtain tail gas content data; comparing the predicted content data with the tail gas content data to obtain a prediction feedback coefficient; adjusting the prediction according to the prediction feedback coefficient.

[0006] Preferably, the gas content in the tail gas is predicted according to the raw material data to obtain predicted content data, comprising: The raw material data is called, wherein the raw material data includes the chemical content of the lead concentrate, the content of the auxiliary material, and the fuel content; The chemical reaction equation of the lead concentrate in the smelting process is constructed, and the greenhouse gas content in the smelting process of the lead concentrate is analyzed by using the law of conservation of mass and the chemical reaction equation to obtain predicted smelting data; wherein the greenhouse gas includes carbon dioxide, carbon monoxide and sulfur dioxide; The auxiliary material content and the fuel content are integrated into a predicted input sequence; the predicted input model is called, and the predicted input sequence is input into the predicted input model to obtain predicted fuel data; The predicted fuel data represents the greenhouse gas content generated by the auxiliary material and the fuel in the smelting process; the predicted input model is constructed based on an artificial intelligence model; The predicted smelting data and the predicted fuel data are summed to obtain predicted gas data; the predicted gas data is adjusted to obtain predicted content data.

[0007] Preferably, the predicted gas data is adjusted to obtain the predicted content data, comprising: The actual measurement value and the measurement variance of the data sensor in a set time period are obtained; the formula The Kalman filter coefficient is calculated; wherein, represents the variance of the model prediction process; represents the measurement variance of the data sensor; The predicted adjustment function is constructed: ; wherein, represents the predicted gas data; represents the actual measurement value; the predicted content data is calculated according to the predicted adjustment function.

[0008] Preferably, the predicted input model is constructed based on an artificial intelligence model, comprising: The appropriate model and deep learning framework are selected from the artificial intelligence model library, and the selected model is constructed by using the deep learning framework to obtain a constructed model; The standard data set is obtained; wherein the standard data set includes standard input data consistent with the content attribute of the predicted input data, and standard output data consistent with the content attribute of the predicted fuel data; The standard data set is divided into a training set, a validation set and a test set according to a set proportion; the constructed model is trained by using the training set; the internal parameters of the constructed model are adjusted by using the validation set; the trained constructed model is tested by using the test set to obtain test indicators; Obtaining an index threshold value, comparing the test index with the index threshold value; when the test index is greater than the index threshold value, the trained construction model is marked as a prediction input model; otherwise, the prediction input model is re-constructed and trained.

[0009] Preferably, the gas content data is fused and analyzed to obtain tail gas content data, including: Accessing gas content data; wherein the gas content data includes: carbon dioxide content, carbon monoxide content and sulfur dioxide content; Integrating the gas content data in the set time period, integrating the gas content data of the same data type, and summing the gas content data of the same data type to obtain the tail gas content data.

[0010] Preferably, the comparison of the predicted content data and the tail gas content data to obtain the adjustment feedback signal includes: Matching the data of the same type in the predicted content data and the tail gas content data; calculating the content difference value of the corresponding tail gas content data and the predicted content data respectively; wherein the content difference value includes: carbon dioxide content difference, carbon monoxide content difference and sulfur dioxide content difference; Comparing the absolute value of the content difference value with the corresponding difference threshold value; when the absolute value of the content difference value is greater than the difference threshold value, the adjustment feedback signal is generated; otherwise, the predicted content data is continuously analyzed.

[0011] Preferably, the difference threshold value is obtained by: Obtaining historical data of a set time period; wherein the historical data includes: historical predicted content data and historical tail gas content data; calculating the difference value of the historical predicted content data and the historical tail gas content data of the same type to obtain a difference value sequence; Calculating the average value and the standard deviation of the difference value sequence respectively; constructing a static threshold function: According to the static threshold function, the static threshold value of the difference value is calculated; wherein, is the average value of the difference value sequence; is the standard deviation of the difference value sequence; represents the tolerance coefficient; Accessing the variance of the model prediction process and the measurement variance of the data sensor; calculating the dynamic threshold value of the difference value through the formula

[0012] Preferably, the adjustment of the prediction according to the adjustment feedback signal includes: ​obtain a feedback adjustment database; when receiving the adjustment feedback signal, match the absolute value of the content difference value with the feedback adjustment database to obtain the corresponding predicted adjustment measure; adjust the prediction process according to the corresponding predicted adjustment measure, and analyze the adjusted prediction process after a set time period.

[0013] It should be noted that the predicted adjustment measure includes: reconstructing the model, retraining the model, calibrating the variance of the model prediction process; calibrating the measurement variance of the data sensor, etc.

[0014] The second aspect of the present application provides a device for accounting for greenhouse gas emissions of a lead smelting plant, comprising: a communication unit and a processing unit; The communication unit is used to obtain raw material data of the lead smelting plant; and detect tail gas in the lead smelting process through a data sensor to obtain gas content data. The processing unit is used to predict the gas content in the tail gas according to the raw material data to obtain predicted content data; perform fusion analysis on the gas content data to obtain tail gas content data; compare the predicted content data with the tail gas content data to obtain a prediction feedback coefficient; and adjust the prediction according to the prediction feedback coefficient.

[0015] The third aspect of the present application provides a computer readable storage medium, which stores instructions, when the instructions run on the device for accounting for greenhouse gas emissions of a lead smelting plant, the device for accounting for greenhouse gas emissions of the lead smelting plant executes the method described in the first aspect and any possible implementation manner of the first aspect.

[0016] Compared with the prior art, the present application has the following advantages: 1. The application retrieves the chemical content, auxiliary material content and fuel content of lead concentrate and other raw material data, and constructs chemical reaction equations based on these data. The greenhouse gas content in the smelting process is analyzed using the law of conservation of elements to obtain predicted smelting data. Considering the composition of the raw materials can more accurately reflect the actual situation in the smelting process, providing a reliable basis for subsequent prediction and analysis; the auxiliary material and fuel content is integrated into the prediction input sequence, the predicted fuel data is obtained through the prediction input model, and then the predicted gas data is obtained by summing the predicted smelting data, and finally the predicted content data is obtained by adjustment. Through the method of integrating different sources of data and comprehensive analysis, the greenhouse gas content produced in the smelting process can be more comprehensively predicted, improving the accuracy and completeness of the prediction; the actual measurement value and measurement variance in the set time period are obtained, the Kalman filtering coefficient is calculated through the formula, and the prediction adjustment function is constructed; the introduction of the Kalman filtering coefficient can comprehensively consider the uncertainty of model prediction and actual measurement, dynamically adjust the prediction result according to the variance of the two, so that the predicted content data is closer to the true value, improving the prediction accuracy and stability; retrieve the gas content data, integrate the data in the set time period, sum the same type of gas content data to obtain the tail gas content data, which can clearly understand the total amount of various greenhouse gases in the tail gas in different time periods, and provide strong data support for subsequent environmental assessment and management.

[0017] 2. The application matches the same type of data in the predicted content data and the tail gas content data, calculates the corresponding content difference value, including the content difference value of carbon dioxide, carbon monoxide and sulfur dioxide, can clearly reflect the deviation between the predicted value and the actual tail gas value, and provides clear data basis for subsequent judgment and adjustment; by comparing the absolute value of the content difference value with the corresponding difference threshold value, it can intelligently judge whether there is a significant difference between the predicted value and the actual value. When the absolute value of the difference is greater than the difference threshold value, an adjustment feedback signal is generated, timely reminding the relevant personnel or system of abnormal conditions, which helps to quickly find problems and take measures to avoid further expansion of the problem; when obtaining the difference threshold value, both historical data and real-time data are considered; by obtaining the historical data of the set time period, the difference sequence of the historical predicted content data and the historical tail gas content data is calculated, and then the static threshold value of the difference is obtained, which reflects the average fluctuation of the data in the long term; at the same time, the variance of the model prediction process and the measurement variance of the data sensor are called, and the dynamic threshold value of the difference is calculated, considering the uncertainty in the current prediction and measurement process. The maximum of the static threshold value and the dynamic threshold value is selected as the difference threshold value, ensuring the scientificity and rationality of the threshold setting, which can more accurately judge whether the predicted value and the actual value are abnormal; the tolerance coefficient n in the static threshold function can be adjusted according to the actual situation, so that the threshold setting has a certain flexibility. The requirements for prediction accuracy may be different under different production scenes or environmental conditions, and by adjusting the n value, different needs can be adapted to, improving the applicability of the whole monitoring and adjustment system; the feedback adjustment database is obtained, and when the adjustment feedback signal is received, the absolute value of the content difference value is matched with the database, which can quickly obtain the corresponding prediction adjustment measures. This database-based matching method makes the adjustment measures targeted and accurate, which can take the most appropriate adjustment strategy according to the specific difference, improving the adjustment efficiency. BRIEF DESCRIPTION OF DRAWINGS

[0018] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, below will briefly introduce the drawings needed to be used in the embodiments or prior art description, obviously, the drawings in the following description are only some embodiments of the present application, and those skilled in the art can obtain other drawings according to these drawings without creating any inventive labor.

[0019] Figure 1 The schematic diagram of the whole steps of the method of the present application; Figure 2 The schematic diagram of the prediction process and actual measurement integration step of the method of the present application; Figure 3 The schematic diagram of the prediction process evaluation step of the present application; Figure 4This is a schematic diagram of the device structure of the present invention. Detailed Implementation

[0020] The technical solution of the present invention will be clearly and completely described below with reference to the embodiments. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0021] Please see Figure 1 The first aspect of this invention provides a method for calculating greenhouse gas emissions from a lead smelter, comprising: S101. Obtain raw material data from the lead smelter; detect the tail gas during the lead smelting process using data sensors to obtain gas content data. S102. Based on the raw material data, predict the gas content in the exhaust gas to obtain the predicted content data; perform fusion analysis on the gas content data to obtain the exhaust gas content data. S103. Compare the predicted content data with the exhaust gas content data to obtain the prediction feedback coefficient; adjust the prediction based on the prediction feedback coefficient.

[0022] Based on the above steps, acquiring raw material data and detecting tail gas content allows for a comprehensive understanding of the actual situation of tail gas emissions from raw material input to production in a lead smelter. This helps in accurately controlling the entire production process and promptly identifying potential problems. Predicting the gas content in tail gas based on raw material data provides a forward-looking reference for the production process, enabling production personnel to understand potential tail gas emissions in advance, facilitating the development of countermeasures and optimizing production operations. Fusion analysis of gas content data yields tail gas content data; compared to single data sources, fusion analysis integrates multiple information sources, reducing errors and improving the accuracy and reliability of tail gas content data. Comparing the predicted content data with the tail gas content data yields a prediction feedback coefficient, which, when used to adjust the prediction, continuously optimizes the prediction model, making the prediction results closer to reality, improving the accuracy and practicality of the prediction, and ultimately enhancing the control level and management efficiency of the entire production process. This contributes to energy conservation and emission reduction, cost reduction, improved product quality, and production safety.

[0023] In one possible implementation of this invention, combined with Figure 1 ,like Figure 2 As shown, the above S102 can be specifically implemented through the following S201-S204, which are explained in detail below: S201, retrieve raw material data and; construct the chemical reaction equation of lead concentrate in the smelting process; use the law of conservation of elements and the chemical reaction equation to analyze the greenhouse gas content in the lead concentrate smelting process, and obtain the predicted smelting data.

[0024] Among them, the raw material data includes: the chemical content of lead concentrate, the content of auxiliary materials and the content of fuel; the greenhouse gas includes: carbon dioxide, carbon monoxide and sulfur dioxide.

[0025] Example: Assuming that a lead smelting plant wants to process a batch of lead concentrate, the retrieved raw material data is as follows: The chemical content of lead concentrate: the main component is lead sulfide (PbS), the content is 80%, and there are a small amount of zinc oxide (ZnO, content 5%), silicon dioxide (Si , content 10%) and other impurities.

[0026] The content of auxiliary materials: limestone (CaC ) is used as auxiliary material, the content is 15 tons, which is used to adjust the slag properties.

[0027] The content of fuel: coke is used as fuel, the fixed carbon content is 85%, and the calorific value is 28 MJ / kg, 20 tons of coke are put into this smelting; Roasting stage: ; Reduction stage: At the same time, incomplete combustion of coke will produce carbon monoxide: ; For sulfur dioxide: according to the content of lead sulfide and the chemical reaction equation, 2 moles of lead sulfide produce 2 moles of sulfur dioxide, and the theoretical amount of sulfur dioxide produced can be calculated. Assuming that the lead sulfide is completely reacted, the molar mass of lead sulfide is 239.3 g / mol, and the molar mass of sulfur dioxide is 64 g / mol, the mass of lead sulfide in the lead concentrate is the total mass multiplied by the content, if the total mass of lead concentrate is 100 tons, the mass of lead sulfide is 80 tons, the mass of sulfur dioxide produced is about 21.4 tons.

[0028] For carbon dioxide (CO2) and carbon monoxide (CO): according to the combustion of coke, assuming the proportion of complete combustion and incomplete combustion, combined with the fixed carbon content in coke and the chemical reaction equation. If it is assumed that 70% of the coke is completely combusted to produce carbon dioxide and 30% is incompletely combusted to produce carbon monoxide, the mass of fixed carbon in coke is 17 tons, according to the stoichiometric relationship between carbon and oxygen, the mass of carbon dioxide produced is about 26.18 tons, and the mass of carbon monoxide produced is about 7.48 tons.

[0029] S202, integrate the auxiliary material content and the fuel content into a prediction input sequence; call a prediction input model, input the prediction input sequence into the prediction input model, obtain prediction fuel data; sum the prediction smelting data and the prediction fuel data to obtain prediction gas data; adjust the prediction gas data to obtain prediction content data.

[0030] Among them, the prediction fuel data represents the greenhouse gas content generated by the auxiliary material and the fuel in the smelting process; the prediction input model is constructed based on an artificial intelligence model.

[0031] In some specific implementations, the prediction input model is constructed based on an artificial intelligence model, including: Filtering a suitable model and a deep learning framework from an artificial intelligence model library, and constructing the filtered model by using the deep learning framework to obtain a constructed model; Obtain a standard data set; wherein the standard data set includes standard input data consistent with the content attributes of the prediction input data, and standard output data consistent with the content attributes of the prediction fuel data; Divide the standard data set into a training set, a validation set and a test set according to a set proportion; use the training set to train the constructed model; use the validation set to adjust the internal parameters of the constructed model; use the test set to test the trained constructed model to obtain a test index; Obtain an index threshold, compare the test index with the index threshold; when the test index is greater than the index threshold, mark the trained constructed model as a prediction input model; otherwise, re-construct and train the prediction input model.

[0032] Example: integrate the auxiliary material content: limestone 15 tons, and the fuel content: coke 20 tons, fixed carbon content 85%, heat value 28MJ / kg, etc. into a prediction input sequence; input the prediction input sequence into the prediction input model to obtain prediction fuel data, assuming that the auxiliary material and the fuel generate 25 tons of carbon dioxide, 7 tons of carbon monoxide, and 20 tons of sulfur dioxide in the smelting process; Sum the prediction smelting data and the prediction fuel data to obtain prediction gas data, the total amount of carbon dioxide is 51.18 tons, the total amount of carbon monoxide is 14.48 tons, and the total amount of sulfur dioxide is 41.4 tons.

[0033] S203, obtain actual measurement values and measurement variances of data sensors in a set time period; calculate the Kalman filtering coefficient by the formula Calculate the Kalman filtering coefficient; construct a prediction adjustment function: ; calculate the prediction content data according to the prediction adjustment function.

[0034] Among them, a variance of a model prediction process; a measurement variance of a data sensor; predicted gas data; actual measurement values.

[0035] Example: Obtain actual measurement values within a set time period of 1 hour, assuming that the actual measurement values of carbon dioxide, carbon monoxide and sulfur dioxide measured by the data sensor are 45 tons, 14 tons and 40 tons respectively; At the same time, obtain the measurement variance of the data sensor, assuming that the measurement variance of carbon dioxide σ_c is 0.5, the measurement variance of carbon monoxide σ_c is 0.3, and the measurement variance of sulfur dioxide σ_c is 0.4; The variance of the model prediction process σ_m is assumed to be 0.3 for carbon dioxide, 0.2 for carbon monoxide, and 0.25 for sulfur dioxide. The Kalman filter coefficient is calculated by the formula, and the carbon dioxide is 0.375; The carbon monoxide is 0.4; The sulfur dioxide is 0.385.

[0036] The predicted content data is calculated according to the prediction adjustment function; wherein, the carbon dioxide is 45.66 tons; The carbon monoxide is 14.16 tons, and the sulfur dioxide is 40.38 tons.

[0037] S204, retrieve gas content data; integrate the gas content data within a set time period, integrate the gas content data of the same data type, and sum the gas content data of the same data type to obtain tail gas content data.

[0038] Among them, the gas content data includes: carbon dioxide content, carbon monoxide content and sulfur dioxide content.

[0039] Example: Retrieve gas content data within a set time period of 1 hour, assuming that the data is as follows: Carbon dioxide tail gas content: 55.66 tons; Carbon monoxide tail gas content: 18.12 tons; Sulfur dioxide tail gas content: 37.28 tons.

[0040] Based on the above steps, the chemical content, auxiliary material content and fuel content of the lead concentrate are called, and the chemical reaction equation is constructed based on these data. The greenhouse gas content in the smelting process is analyzed by using the law of conservation of elements, and the predicted smelting data is obtained. This way of comprehensively considering the composition of raw materials can more accurately reflect the actual situation in the smelting process, and provide a reliable basis for subsequent prediction and analysis; the auxiliary material and fuel content is integrated into the prediction input sequence, and the predicted fuel data is obtained through the prediction input model, and then the predicted gas data is obtained by summing the predicted smelting data, and finally the predicted content data is obtained. This method of integrating data from different sources and comprehensive analysis can more comprehensively predict the greenhouse gas content generated in the smelting process, improve the accuracy and completeness of the prediction; the actual measurement value and measurement variance in the set time period are obtained, the Kalman filtering coefficient is calculated by formula, and the prediction adjustment function is constructed. The introduction of Kalman filtering coefficient can comprehensively consider the uncertainty of model prediction and actual measurement, dynamically adjust the prediction result according to the variance of the two, so that the predicted content data is closer to the true value, and the prediction accuracy and stability are improved; the gas content data is called, and the data in the set time period is integrated. The same type of gas content data is summed to obtain the tail gas content data. This integration method can clearly understand the total amount of various greenhouse gases in the tail gas in different time periods, and provide strong data support for subsequent environmental assessment and management.

[0041] In a possible implementation manner of the embodiment of the present application, in combination with Figure 1 As shown in the figure, the above S103 can be implemented by the following S301-S304, which will be described in detail as follows. Figure 3 S301, match the data of the same type in the predicted content data and the tail gas content data; calculate the content difference value of the corresponding tail gas content data and the predicted content data respectively.

[0042] Among them, the content difference value includes: carbon dioxide content difference value, carbon monoxide content difference value and sulfur dioxide content difference value.

[0043] Example: the predicted content data is: carbon dioxide is 45.66 tons; carbon monoxide is 14.16 tons, and sulfur dioxide is 40.38 tons; The tail gas content data is: carbon dioxide is 55.66 tons; carbon monoxide is 18.12 tons; sulfur dioxide is 37.28 tons.

[0044] The absolute value of the calculated content difference value is: carbon dioxide is 10 tons; carbon monoxide is 3.96 tons; sulfur dioxide is 3.1 tons.

[0045] ​S302, compare the absolute value of the content difference value with the corresponding difference value threshold; when the absolute value of the content difference value is greater than the difference value threshold, an adjustment feedback signal is generated; otherwise, the predicted content data is continuously analyzed.

[0046] In some specific implementations, the difference value threshold is obtained in the following manner: Obtain historical data of a set time period; wherein the historical data includes: historical predicted content data and historical tail gas content data; calculate the difference value of the historical predicted content data and the historical tail gas content data of the same data type respectively to obtain a difference value sequence; Calculate the average value and the standard deviation of the difference value sequence respectively; construct a static threshold function: Calculate the static threshold value of the difference value according to the static threshold function; wherein, is the average value of the difference value sequence; is the standard deviation of the difference value sequence; represents the tolerance coefficient; Obtain the variance of the model prediction process and the measurement variance of the data sensor; calculate the dynamic threshold value of the difference value through the formula Select the maximum value from the static threshold value and the dynamic threshold value as the difference value threshold.

[0047] Example: obtain the historical prediction and actual measurement value of each hour within the past 24 hours to obtain the historical prediction and actual measurement C Difference value sequence: [+10, -5, -3, +8, +2,...]

[0048] Calculate the average value of the difference value sequence μ≈0.67ppm; the standard deviation σ≈6.65ppm; calculate the static threshold value through the formula: 13.97; The variance of the model prediction process and the measurement variance of the data sensor are 0.5 and 0.3 respectively; calculate the dynamic threshold value through the formula: 1.6; select 13.97 as the difference value threshold; Similarly, the difference value threshold of carbon monoxide is 8; the difference value threshold of sulfur dioxide is 3; the absolute value of the content difference value of sulfur dioxide is greater than the difference value threshold, so a sulfur dioxide adjustment signal is generated.

[0049] S303, obtain a feedback adjustment database; when the adjustment feedback signal is received, match the absolute value of the content difference value with the feedback adjustment database to obtain the corresponding predicted adjustment measure.

[0050] S304, adjust the prediction process according to the corresponding predicted adjustment measure, and analyze the adjusted prediction process after a set time period.

[0051] The absolute value of the content difference value of sulfur dioxide is matched with the feedback adjustment database to obtain a corresponding adjustment measure, i.e., changing the calibration frequency of the data sensor from once a week to once every three days.

[0052] Based on the above steps, the same type of data in the predicted content data and the tail gas content data is matched, and the corresponding content difference values, including the content difference values of carbon dioxide, carbon monoxide and sulfur dioxide, are calculated. This accurate quantification method can clearly reflect the deviation between the predicted value and the actual tail gas value, providing clear data basis for subsequent judgment and adjustment; by comparing the absolute value of the content difference value with the corresponding difference threshold value, it can be intelligently judged whether there is a significant difference between the predicted value and the actual value. When the absolute value of the difference is greater than the difference threshold value, an adjustment feedback signal is generated to timely remind the relevant personnel or system of abnormal conditions, which helps to quickly find problems and take measures to avoid further expansion of the problem; when obtaining the difference threshold value, both historical data and real-time data are considered. By obtaining the historical data of a set time period, the difference sequence of the historical predicted content data and the historical tail gas content data is calculated, and then the static threshold value of the difference is obtained, which reflects the average fluctuation of the data in the long term. At the same time, the variance of the model prediction process and the measurement variance of the data sensor are retrieved, and the dynamic threshold value of the difference is calculated, considering the uncertainty in the current prediction and measurement process. The maximum of the static threshold value and the dynamic threshold value is selected as the difference threshold value, ensuring the scientificity and rationality of the threshold setting, which can more accurately judge whether the predicted value and the actual value are abnormal; the tolerance coefficient n in the static threshold function can be adjusted according to the actual situation, so that the threshold setting has a certain flexibility. In different production scenes or environmental conditions, the requirement for prediction accuracy may be different. By adjusting the n value, different needs can be adapted to, improving the applicability of the whole monitoring and adjustment system; the feedback adjustment database is obtained, and when the adjustment feedback signal is received, the absolute value of the content difference value is matched with the database, which can quickly obtain the corresponding prediction adjustment measure. This database-based matching method makes the adjustment measure targeted and accurate, which can take the most appropriate adjustment strategy according to the specific difference, improving the adjustment efficiency.

[0053] Referring to Figure 4 The second aspect of the present application provides a device for calculating the greenhouse gas emissions of a lead smelting plant, comprising a communication unit and a processing unit. The communication unit is used to obtain raw material data of the lead smelting plant; and a data sensor is used to detect the tail gas in the lead smelting process to obtain gas content data. The processing unit is configured to predict the gas content in the tail gas according to the raw material data, obtain predicted content data, perform fusion analysis on the gas content data, obtain tail gas content data, compare the predicted content data with the tail gas content data, and obtain a prediction feedback coefficient; and adjust the prediction according to the prediction feedback coefficient.

[0054] The device can also include a storage unit that stores program codes and data of the accounting device for greenhouse gas emissions of the lead smelting plant.

[0055] The embodiment of the present application can divide the circuit auxiliary market operation strategy optimization device based on model analysis into functional units according to the above-mentioned method examples. For example, each functional unit can be divided according to each function, or two or more functions can be integrated into one processing unit. The integrated unit can be realized in the form of hardware or software functional unit. It should be noted that the division of units in the embodiment of the present application is illustrative, and is only a logical functional division. In actual implementation, there can be another division method.

[0056] The third aspect of the embodiment of the present application provides a computer readable storage medium, and the computer readable storage medium stores instructions. When the instructions run on the accounting device for greenhouse gas emissions of the lead smelting plant, the accounting device for greenhouse gas emissions of the lead smelting plant executes the method described in the first aspect and any possible implementation manner of the first aspect.

[0057] Some data in the above formula are calculated by removing the dimension and taking the numerical value. The formula is obtained by software simulation of a large amount of collected data to obtain a formula closest to the real situation. The preset parameters and the preset threshold in the formula are set by a person skilled in the art according to the actual situation or obtained by a large amount of data simulation.

[0058] The working principle of the present application is as follows: raw material data of a lead smelting plant is obtained; tail gas in the lead smelting process is detected by a data sensor to obtain gas content data; the gas content in the tail gas is predicted according to the raw material data to obtain predicted content data; fusion analysis is performed on the gas content data to obtain tail gas content data; the predicted content data is compared with the tail gas content data to obtain a prediction feedback coefficient; and the prediction is adjusted according to the prediction feedback coefficient.

[0059] The above embodiments are only used to illustrate the technical method of the present application and are not limiting. Although the present application has been described in detail with reference to the preferred embodiments, it should be understood by those skilled in the art that the technical method of the present application can be modified or replaced equivalently without departing from the spirit and scope of the technical method of the present application.

Claims

1. A method for calculating greenhouse gas emissions from lead smelting plants, characterized in that, include: Obtain raw material data from lead smelters; use data sensors to detect tail gas during lead smelting to obtain gas content data; The gas content in the exhaust gas is predicted based on the raw material data, and the predicted content data is obtained. By fusing and analyzing the gas content data, exhaust gas content data can be obtained. The predicted content data is compared with the exhaust gas content data to obtain the prediction feedback coefficient; The forecast is adjusted based on the forecast feedback coefficient.

2. The method for calculating greenhouse gas emissions from a lead smelter according to claim 1, characterized in that, The step of predicting the gas content in the exhaust gas based on raw material data to obtain predicted content data includes: Retrieve raw material data; the raw material data includes: the chemical content of lead concentrate, the content of auxiliary materials, and the fuel content; Chemical reaction equations for lead concentrate smelting were constructed; the greenhouse gas content during lead concentrate smelting was analyzed using the law of conservation of elements and chemical reaction equations to obtain predicted smelting data; among which, greenhouse gases include: carbon dioxide, carbon monoxide and sulfur dioxide. The auxiliary material content and fuel content are integrated into a prediction input sequence; the prediction input model is called, and the prediction input sequence is input into the prediction input model to obtain the prediction fuel data; Among them, the predicted fuel data represents the greenhouse gas content generated by auxiliary materials and fuels during the smelting process; the predicted input model is built based on an artificial intelligence model; The predicted smelting data and the predicted fuel data are summed to obtain the predicted gas data; the predicted gas data are then adjusted to obtain the predicted content data.

3. The method for calculating greenhouse gas emissions from a lead smelter according to claim 2, characterized in that, The process of adjusting the predicted gas data to obtain the predicted content data includes: Obtain the actual measured values ​​and the measurement variance of the data sensor within a set time period; using the formula... Calculate the Kalman filter coefficients; where, This represents the variance in the model's prediction process; This represents the measurement variance of the data sensor; Construct the prediction adjustment function: ;in, This indicates predicted gas data; This represents the actual measured value; the predicted content data is calculated based on the prediction adjustment function.

4. The method for calculating greenhouse gas emissions from a lead smelter according to claim 2, characterized in that, The predicted input model is built based on an artificial intelligence model and includes: Select suitable models and deep learning frameworks from the artificial intelligence model library, and use the deep learning framework to build the selected models to obtain the constructed models; Obtain the standard dataset; wherein, the standard dataset includes standard input data consistent with the content attributes of the prediction input data, and standard output data consistent with the content attributes of the prediction fuel data; The standard dataset is divided into a training set, a validation set, and a test set according to a set ratio; the model is trained using the training set; the internal parameters of the model are adjusted using the validation set; and the trained model is tested using the test set to obtain test metrics. Obtain the indicator threshold and compare the test indicator with the indicator threshold; when all test indicators are greater than the indicator threshold, mark the trained model as the prediction input model; otherwise, retrain the prediction input model.

5. The method for calculating greenhouse gas emissions from a lead smelting plant according to claim 1, characterized in that, The process of fusing and analyzing gas content data to obtain exhaust gas content data includes: Retrieve gas content data; the gas content data includes: carbon dioxide content, carbon monoxide content, and sulfur dioxide content; The gas content data within a set time period are integrated, gas content data of the same data type are integrated, and the gas content data of the same data type are summed to obtain the exhaust gas content data.

6. The method for calculating greenhouse gas emissions from a lead smelter according to claim 1, characterized in that, The step of comparing the predicted content data with the exhaust gas content data to obtain the adjustment feedback signal includes: Match data of the same data type in the predicted content data and the exhaust gas content data; calculate the content difference between the corresponding exhaust gas content data and the predicted content data; the content difference includes: carbon dioxide content difference, carbon monoxide content difference, and sulfur dioxide content difference; The absolute value of the content difference is compared with the corresponding difference threshold; when the absolute value of the content difference is greater than the difference threshold, an adjustment feedback signal is generated; otherwise, the predicted content data is continuously analyzed.

7. The method for calculating greenhouse gas emissions from a lead smelter according to claim 6, characterized in that, The method for obtaining the difference threshold includes: Acquire historical data for a specified time period; the historical data includes historical predicted content data and historical exhaust gas content data; calculate the difference between the historical predicted content data and the historical exhaust gas content data, which are of the same data type, to obtain a difference sequence; Calculate the mean and standard deviation of the difference sequence respectively; construct a static threshold function: The static threshold of the difference is calculated based on the static threshold function; where, The average of the difference sequence; The standard deviation of the difference sequence; This represents the tolerance coefficient; Retrieve the variance of the model prediction process and the measurement variance of the data sensors; using the formula Calculate the dynamic threshold of the difference; select the larger of the static and dynamic thresholds as the difference threshold.

8. The method for calculating greenhouse gas emissions from a lead smelter according to claim 1, characterized in that, The step of adjusting the prediction based on the adjustment feedback signal includes: Obtain feedback adjustment database; when adjustment feedback signal is received, match the absolute value of the content difference with the feedback adjustment database to obtain the corresponding predictive adjustment measures; The forecasting process is adjusted according to the corresponding forecasting adjustment measures, and the adjusted forecasting process is analyzed and feedback is provided after a set time period.

9. A device for calculating greenhouse gas emissions from a lead smelter, applicable to the method for calculating greenhouse gas emissions from a lead smelter as described in any one of claims 1-8, characterized in that, include: Communication unit and processing unit; The communication unit is used to acquire raw material data from the lead smelter; and to detect the exhaust gas during the lead smelting process using a data sensor to obtain gas content data. The processing unit is used to predict the gas content in the exhaust gas based on the raw material data to obtain predicted content data; to perform fusion analysis on the gas content data to obtain exhaust gas content data; to compare the predicted content data with the exhaust gas content data to obtain a prediction feedback coefficient; and to adjust the prediction based on the prediction feedback coefficient.

10. A computer-readable storage medium storing instructions that, when executed on a greenhouse gas emissions accounting device for a lead smelter, cause the greenhouse gas emissions accounting device for a lead smelter to perform the method according to any one of claims 1-8.

Images